Cellular and molecular characterization of the immune pathogenesis of insulin-dependent diabetes mellitus (IDDM) is necessary for the rational design of specific therapies for the prevention and treatment of diabetes. However, successful immunotherapy for the prevention of IDDM mandates identification of the relevant target antigen(s) involved in the autoimmune process. This current proposal seeks to explore several novel therapeutic strategies for the prevention and treatment of autoimmune diabetes in the NOD mouse. There are three major aims of this proposal: First, the non-obese diabetic (NOD) murine model of IDDM will be longitudinally probed for a spontaneous loss of tolerance to two major putative autoantigens at a molecular level. We will utilize overlapping glutamic acid decarboxylase-65 (GAD) and insulin peptides to determine the immunogenicity and specificity of islet-derived T cells. Molecular analysis of cytokine mRNA transcript levels elaborated by T cells responsive to autoantigens will elicit definitive data regarding the T helper subset requisite for the onset and progression of autoimmune diabetes. Furthermore, we propose to evaluate the diabetogenic potential of GAD and insulin peptide-reactive islet-derived T cells by performing adoptive transfer experiments to NOD scid/scid recipients. Finally, we propose to use the novel technique of gene therapy to express immunomodulatory proteins which can ameliorate the vulnerability of islets to recurrent autoimmune destruction. In specific aim 2, we propose to examine the potential of intrathymic inocula of bone marrow and islet cells for "rescue" from the destruction of residual pancreatic b-cells in the NOD mouse. Moreover, these inocula will also e used to promote donor-specific unresponsiveness to extrathymic islet allografts in long-term diabetic NOD mice. Furthermore, we propose to develop a large animal model for the preclinical evaluation of the potential of thymus-mediated immune tolerance in man. In specific aim 3, the mechanisms of leukocyte transendothelial migration and their role in the recruitment of leukocytes to the pancreatic islet milieu will be studied. We propose to assess the potential of blocking platelet endothelial cell adhesion molecule-1 (PECAM-1) to treat the diabetogenic process. We also seek to examine the mechanisms of CD4- mediated signal transduction and its role in the activation of autoantigen-specific T cells. By utilizing a novel CD4 cyclic analogue, we seek to both treat and prevent the onset of autoimmune diabetes by blocking autoimmune responses in an antigen-specific manner. Furthermore, we will utilize the information obtained from Specific Aim 1 to create novel glycosylated peptides which have increased stability and the potential to suppress autoantigen-reactive T cells in an peptide-specific manner.